napsternxg
commited on
Commit
•
16cff4f
1
Parent(s):
4f3ddfe
Update README.md
Browse files
README.md
CHANGED
@@ -94,4 +94,70 @@ size_categories:
|
|
94 |
* query_class - category to which the query falls under
|
95 |
* Annotated (product,relevance judgement) pairs, columns:
|
96 |
* id - Unique ID for each annotation
|
97 |
-
* label - Relevance label, one of 'Exact', 'Partial', or 'Irrelevant'
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
94 |
* query_class - category to which the query falls under
|
95 |
* Annotated (product,relevance judgement) pairs, columns:
|
96 |
* id - Unique ID for each annotation
|
97 |
+
* label - Relevance label, one of 'Exact', 'Partial', or 'Irrelevant'
|
98 |
+
|
99 |
+
# Citation
|
100 |
+
|
101 |
+
Please cite this paper if you are building on top of or using this dataset:
|
102 |
+
|
103 |
+
```text
|
104 |
+
@InProceedings{wands,
|
105 |
+
title = {WANDS: Dataset for Product Search Relevance Assessment},
|
106 |
+
author = {Chen, Yan and Liu, Shujian and Liu, Zheng and Sun, Weiyi and Baltrunas, Linas and Schroeder, Benjamin},
|
107 |
+
booktitle = {Proceedings of the 44th European Conference on Information Retrieval},
|
108 |
+
year = {2022},
|
109 |
+
numpages = {12}
|
110 |
+
}
|
111 |
+
```
|
112 |
+
|
113 |
+
|
114 |
+
# Code for generating dataset
|
115 |
+
|
116 |
+
|
117 |
+
```python
|
118 |
+
import pandas as pd
|
119 |
+
from datasets import Dataset
|
120 |
+
from datasets import DatasetDict, Dataset
|
121 |
+
from datasets import ClassLabel, load_from_disk, load_dataset, concatenate_datasets
|
122 |
+
from pathlib import Path
|
123 |
+
|
124 |
+
base_path = "https://github.com/wayfair/WANDS/raw/main/dataset/"
|
125 |
+
|
126 |
+
query_df = pd.read_csv(f"{base_path}/query.csv", sep='\t')
|
127 |
+
product_df = pd.read_csv(f"{base_path}/product.csv", sep='\t')
|
128 |
+
label_df = pd.read_csv(f"{base_path}/label.csv", sep='\t')
|
129 |
+
|
130 |
+
df_dataset = label_df.merge(
|
131 |
+
query_df, on="query_id"
|
132 |
+
).merge(
|
133 |
+
product_df, on="product_id"
|
134 |
+
)
|
135 |
+
|
136 |
+
wands_class_label_feature = ClassLabel(num_classes=3, names=["Irrelevant", "Partial", "Exact"])
|
137 |
+
dataset = dataset.train_test_split(test_size=2/5, seed=1337)
|
138 |
+
dev_test_dataset = dataset["test"].train_test_split(test_size=1/2, seed=1337)
|
139 |
+
dataset = DatasetDict(
|
140 |
+
train=dataset["train"],
|
141 |
+
dev=dev_test_dataset["train"],
|
142 |
+
test=dev_test_dataset["test"],
|
143 |
+
)
|
144 |
+
"""
|
145 |
+
DatasetDict({
|
146 |
+
train: Dataset({
|
147 |
+
features: ['id', 'query_id', 'product_id', 'label', 'query', 'query_class', 'product_name', 'product_class', 'category hierarchy', 'product_description', 'product_features', 'rating_count', 'average_rating', 'review_count'],
|
148 |
+
num_rows: 140068
|
149 |
+
})
|
150 |
+
dev: Dataset({
|
151 |
+
features: ['id', 'query_id', 'product_id', 'label', 'query', 'query_class', 'product_name', 'product_class', 'category hierarchy', 'product_description', 'product_features', 'rating_count', 'average_rating', 'review_count'],
|
152 |
+
num_rows: 46690
|
153 |
+
})
|
154 |
+
test: Dataset({
|
155 |
+
features: ['id', 'query_id', 'product_id', 'label', 'query', 'query_class', 'product_name', 'product_class', 'category hierarchy', 'product_description', 'product_features', 'rating_count', 'average_rating', 'review_count'],
|
156 |
+
num_rows: 46690
|
157 |
+
})
|
158 |
+
})
|
159 |
+
"""
|
160 |
+
|
161 |
+
dataset.push_to_hub("napsternxg/wands")
|
162 |
+
|
163 |
+
```
|